In this paper, an adaptive high-accurate observer-based fault detection approach for industrial applications is proposed. The proposed fault detection algorithm employs a fuzzy logic-based approach with the objective of finding the appropriate observer gains that could cope with the different working conditions. The flexibility and adaptability represent the main objectives of the proposed observer. This work is interested in proposing an observer with fuzzy variable gains for a general nonlinear system. Furthermore, a linear model has been built to facilitate the accomplishment of the fault detection of the industrial servo system by using the proposed observer. In order to evaluate the proposed approach, eleven realistic sensor fault scenarios are created under varying conditions: fault parameters (e.g., multiple fault profiles, location, and magnitudes), unknown inputs (e.g., disturbers and sensor noises) for performance testing. Also, a scoring algorithm has been implemented, to evaluate the classification ability of the algorithm and the early fault detection ability. The experimental results demonstrate the effectiveness of the proposed observer approach in detecting sensor faults in the industrial servo systems, with 88.8% classification accuracy. Furthermore, the obtained results confirm the proposed algorithm superiority when compared to classical Luenberger observer with constant gains.

The conventional methods of observer poles placement in sensor fault detection usually adopt the trial-and-error methods. These methods cannot achieve global optimal performance because of their fixed poles placement and it leads to an observer with constant parameters, which could be reducing the system performance. Therefore, this paper proposes a fuzzy-based observer tuning method to optimize and adapt the selection of poles locations to determine the optimal gains of the observer, and it is experimentally applied to a composite sensor fault detection. Fuzzy logic is a promising method that could overcome the trial-and-error method challenges by introducing better adaptation and system robustness. The proposed observer structure includes adaptive tuning corresponding to an unknown input. Utilizing self-tuning for the observer correction stage, the gain is going to be updated online using the proposed fuzzy adaptive poles placement (FAPP) system. This paper validated the system simulation by implementing fault detection algorithms by using a real-time embedded observer-based system. The experimental results demonstrate the effectiveness of the proposed fuzzy-based observer schemes at detecting sensor faults in the Brushless DC (BLDC) motors, with significantly better performance than conventional counterparts’ methodologies. The experiments indicate that the average estimation error is 0.146, which less by 43.8% than was obtained for high levels of noise and disturbances compared with the traditional Luenberger observer approach.

Fault detection (FD) techniques play a crucial role at increasing the reliability, maintainability and safety for engineering applications. Meanwhile, Brushless Direct Current (BLDC) motor is a main integrated part for engineering systems in several mechatronics and robotics applications. Therefore, this work presents a model-based sensor FD approach. The proposed approach is tailored for detecting speed sensor faults at BLDC motor through exploiting Luenberger observer. Several experiments are conducted in order to validate the performance of the proposed approach. During the conducted experiments different faults types are applied to the motor in order to test the proposed FD approach. The experimental results demonstrate the ability of the proposed approach to detect the time and size of the sensor faults.

Fault detection and diagnosis (FDD) are important engineering tasks that could effectively improve safety and reliability of technical processes by reducing the number of shutdowns that are necessary for traditional maintenance routines. Many techniques have been proposed in the area of FDD and the use of observers in fault detection is well established. This paper presents an improvement for Luenberger observer using fuzzy logic for detecting sensor faults. The proposed observer system is characterized by its simple design where the location of the poles of the observer is determined through exploiting a fuzzy logic approach and the gain could be adapted according to the system sates. Extensive experiments are conducted in order to investigate the effectiveness of the proposed approach. Several faults categories ranging from simple faults to complex faults are employed. The experimental results demonstrate that the improved fuzzy Luenberger observer is more effective for fault detection purposes when compared with the classical design approach for Luenberger observer.

Industrial systems serve us in all areas of life. Faults may result in economic loss and wasting energy. Detecting the onset of faults, and determining their location are important engineering tasks. An important class of fault detection (FD) and diagnosis methods utilizes the mathematical model of the monitored system. But, the parameters required for mathematical modelling are limited or unavailable for the most real industrial engineering applications. Observer-based FD is one of the main approaches to FD and identification. At the same time, the traditional observer’s gain calculation required system model parameters. So, this article presents the design of a novel observer for FD purposes using the input–output measurements of the system with unknown parameters. This proposed observer’s design considers observer’s gain tuning, regardless of the mathematical representation of the plant. This the new feature that distinction our observer will facilitate the implementation of FD systems for many unknown parameters industrial systems. The effectiveness of the proposed observer is verified by experimental application to BLDC motor and compared with classical Luenberger observer. The experimental and comparison results prove feasibility and effectiveness of the proposed observer for FD purposes.